Qwopus3-5-9B-Coder-Flash-oQ4e-mtp

This is an oQ4e MLX conversion produced with oMLX.

Source and credit

• Original local source checkpoint: Qwopus3-5-9B-Coder-Flash-bf16
• Source repository / credit: https://huggingface.co/Jackrong/Qwopus3-5-9B-Coder-Flash

Please follow the original model's license, acceptable-use policy, and attribution requirements.

Conversion

• Quantization: oQ4e (enhanced=True)
• Importance calibration: imatrix, 128 samples × 512 tokens
• oMLX version: 0.6.3rc2

oQe uses an imatrix sensitivity calibration to assign precision selectively, rather than applying a uniform bit-width to every tensor. The generated oq_imatrix_report.json records the resulting quantization allocation.

python run_oqe_conversion.py Qwopus3-5-9B-Coder-Flash-bf16 --models-dir /path/to/models --oq-level 4.0 --imatrix-samples 128 --imatrix-seq-length 512 --source-repo https://huggingface.co/Jackrong/Qwopus3-5-9B-Coder-Flash

Files

• model.safetensors or model-*-of-*.safetensors: quantized MLX weights
• model.safetensors.index.json: shard index (present for sharded outputs)
• config.json: model configuration
• oq_imatrix_report.json: imatrix calibration and quantization report